一种基于模糊排队论的网络拥塞控制算法  

A Network Congestion Control Algorithm Based on Fuzzy Queuing Theory

作　　者：张永林[1] 令晓明[1] 

机构地区：[1]兰州交通大学光电技术与智能控制教育部重点实验室,甘肃兰州730070

出　　处：《科技信息》2013年第11期100-101,138,共3页Science & Technology Information

摘　　要：在分析和比较现有的主动队列管理(AQM)的网络拥塞控制算法基础上,将模糊控制理论与排队论相结合,提出了一种适合于动态控制队列长度的拥塞控制算法。该算法根据路由器中队列长度的变化情况,对源端数据包的丢弃概率进行模糊控制,从而避免网络拥塞,提高路由器处理的实时性。通过matlab仿真,验证了此算法能够减小排队的延时、提高带宽利用率以及稳定队列的长度。Analyzing and comparing the network congestion control algorithm based on existing Active Queue Management （AQM）.Combining fuzzy control theory with queuing theory knowledge,congestion control algorithm which is suitable for dynamic control queue length.The algorithm is based on the router queue length changes,fuzzy control source-side data packet drop probability,in order to avoid network congestion,and improve the real-time processing of router. According to Matlab simulation,which shows that this algorithm can reduce the queuing delay, improve bandwidth utilization and stable queue length.

关 键 词：模糊控制 排队论 平均等待时间 队列长度 

分 类 号：TP393.4[自动化与计算机技术—计算机应用技术]